JPS6072330A - Automatic disabler for echo controlling device - Google Patents

Abstract

PURPOSE:To stop the operation of an echo controlling device, by providing a frequency analyzing circuit, deciding circuit, and control circuit and detecting the non-sound signal of an MODEM, facsimile, etc., which do not send any conditioning tones. CONSTITUTION:An input signal from an input terminal is divided into four bands of 0-850Hz, 850-1,700Hz, 1700-2,550Hz, and >=2,550Hz by filters 21, 22-24 of a frequency analyzing section 2 and signals of each band are measured in terms of the level by signal level measuring instruments 25, 26-28. These measured outputs are sent to a deciding section 3 where the intensity of signals of each band is compared and their order is determined by means of a comparator circuit 31. Then a deciding circuit 32 decides whether an input signal is a sound signal or non-sound signal in accordance with a previously stored deciding criterion. A control circuit 4 detects the non-sound signal of the MODEM, facsimile, etc., which do not send any conditioning tones by using the decided result and can stop the operation of an echo controlling device.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an automatic isolater for an echo control device that controls the operation of the echo control device to eliminate the effects of echo in a communication line.

In recent years, due to the rapid increase in long-distance lines, especially satellite communications, the existence of echoes occurring on a single communication line has become a serious problem in terms of speech quality due to the large delay of the line. For this reason, echo control devices such as an echo control device 1 and an echo canceller collar have been conventionally used to eliminate the influence of echoes. But the modem
In non-voice communications such as facsimiles, modems and facsimiles used in international communications use echo control devices to prevent echo control devices from changing communication channel characteristics and causing data errors. Non-voice communications such as these were sending out conditioning tones prescribed by CCITT and stopping the operation of the echo control equipment.

On the other hand, in domestic communications, echo control devices are not used because the delay per line is small, and modems, facsimiles, etc. that do not send out conditioning tones are also used.

However, in recent years, due to the use of domestic satellites and the provision of new services, two-line delay has increased and the need for a single-reverberation control device has increased. The need for deactivation of echo control equipment has been questioned.

The present invention is directed to non-voice communications such as modems and facsimiles that do not transmit conditioning tones.

In order to achieve the above object, the present invention analyzes the frequency spectrum of an input signal to stop the operation of the echo control device.

According to the present invention, there is provided a frequency analysis circuit that performs frequency analysis of an input signal, a determination circuit that determines whether the input signal is a non-speech signal based on the output of the frequency analysis circuit, and a determination result of the first determination circuit. An automatic disabler for a reverberation control device is obtained, which is composed of a control circuit that controls the operation of the control device, and the judgment in the above device is that the vowels that make up the majority of Japanese speech have a fundamental frequency of about 100 to 300 Hz. and its harmonics, and 50% of the
is below 300 Hz, and 80% is below 900 Hz, whereas data, facsimile, etc. are generally below 1
Since carrier waves of kHz or higher are modulated, the power is concentrated at the carrier frequency and the i+ll band waves that occur before and after the carrier frequency, and while voice still generates many harmonics relative to the fundamental wave, data, Even in FM modulation and PM modulation, which generate many sidebands in facsimile machines, the main sidebands are up to about the fourth order. Therefore, by looking at the distribution of the signal spectrum or the number of harmonics and sidebands, it is possible to determine whether the audio signal is This is done by taking advantage of the ability to distinguish between non-audio signals and non-audio signals.

Next, details a1 will be explained with reference to the drawings.

FIG. 1 is a block diagram showing the basic configuration of an automatic disabler for an echo control device according to the present invention.

Input terminal 1 is connected to a communication line. The signal level in each frequency band of the signal inputted from the input terminal is detected by the frequency analysis section 2. In the determination section 3, the frequency analysis section 2
Based on the output of the input signal, it is determined whether the input signal is an audio signal or a non-audio signal. The control circuit 4 converts the output of the determination unit 3 into a signal format that can control the operation of the echo control device. This converted control signal is outputted from an output terminal 5 to control the operation of a reverberation control device (not shown).

As for the structure of the frequency analysis section 2, simple frequency analysis can be performed by using a filter group.

As will be explained later, the configuration of the determination unit 3 includes a method in which the spectral power of the input signal is used as a determination criterion, and a method in which the number of high frequencies or sideband waves is used as a determination criterion.

Next, before going into a more specific explanation of the automatic disabler for an echo control device according to the present invention, the general spectrum of audio and modulated waves will be explained.

FIG. 2 is a diagram showing the normal shape of the spectrum of voice and modulated waves, in which (a) is the spectrum of the voice, (b)
) is a data signal rate of 1200 baud, a data signal rate of 24
Spectrum of a four-phase phase modulated wave when the line frequency is 1800 Hz or less, (C) is the data signal rate of 1200 baud, data signal rate of 1200 bits/second or less, line frequency F7 is 13.00 Hz, FA is 2
The spectrum of the frequency modulated wave in the case of 100Hz is shown respectively. As is clear from Fig. 2, according to (a), the power of the audio spectrum is 5
Up to 0% is below 300 Hz, up to 75% is 900 Hz
According to (b) and (c), the data signal has a carrier wave above 1 kHz, whereas it concentrates below Hz.
The spectrum is also concentrated above 1 kHz. Note that Fig. 1 only shows two examples (b) and (e) as data signals, but these are F in other normal examples.
is 1100Hz + FA is 1300Hz, or F2 is 980. :Although the spectra explained above are at a single point in time, some of these spectra change over time and are difficult to change. Each type also has its own characteristics regarding the number of harmonics or sidebands. Therefore, the specific examples will differ depending on which special Myanmar is used as the criterion.

FIG. 3 is a diagram illustrating the configuration of one embodiment of the automatic disabler according to the present invention shown in FIG. It is a device that makes a judgment by comparing the power of Third
In the figures, reference numerals 1 to 4 refer to the same items as in FIG. The signal input from input terminal 1 is 0~850H3, 850~1 in frequency analyzer 2.
700 Hz, 1700-2500 Hz and 2
Filter 21.2 that passes each band of 550Hz or higher
2. .

23, and 24 into four bands.

The signal level of each divided band is measured by signal level measuring circuits 25, 26, 27' and 28 in the frequency analyzer 2 as well. Each measured level output is sent to the determination section 3, and the comparison circuit 31 compares the strength of the sent signals in each band and ranks the signal strengths. Judgment circuit 3
2, the criterion for determining whether the input signal is an audio signal or a non-audio output is stored in advance for the pattern of the order of the strength of the input signal, and when the ranked signal is input, the criterion is stored. It is determined whether the input signal is audio or not according to the following.

The judgment criteria for the above are determined based on FIG. 2 and the explanation given in connection with FIG. 2, and are audio:
If so, the output of the signal level measuring circuit 25 becomes the first strength point, and the output of the signal level measuring circuit 26, 27 or 28 becomes the second strength point. Third. Or become the fourth strength. On the other hand, for non-audio signals, the output of signal line 15 is the fourth point and the weakest point. In this way, the entire pattern of output from the comparator circuit 31 is outputted through the control interface circuit 34. In the above embodiment, the comparator 31 is used to compress the signal level information, but the 1 determination circuit 32
If there is sufficient capacity, the outputs of the signal level measuring circuits 25 to 28 may be directly input to the determination circuit 32.

FIG. 4 is a diagram showing the structure of another specific example of the automatic disabler according to the present invention shown in FIG. 1. The criteria for this device is that when looking at the short-time spectrum of the input signal, many spectrum peaks (points where the spectrum power is larger than the previous and previous frequencies) occur because audio is a harmonic of the fundamental frequency. On the other hand, in a data signal, sideband waves are generated by modulating a carrier wave, so the number of sideband waves is limited. In Figure 4,
The input signal is divided into each band of the frequency range shown in the diagram by eight filters (collectively indicated by 41, the same applies hereinafter), and the signal level is measured by a signal level measuring device 42.

However, when the measured value is below a predetermined threshold, it is set to 0 for operational stability reasons. Note that the filter 41 and the signal level measuring device 42 form the frequency analysis section 2 in FIG. The comparison circuit 43 compares the signal of each band to see if the output of the measuring device 42 is larger or smaller than the bands before and after it, and if it is larger, outputs +1111.

This It I II output indicates that the B input is larger than the A output and the C output, that is, it corresponds to the peak of the spectrum. The adder 44 adds the outputs of all comparison circuits 43 to calculate the number of peaks.

A comparator 45 compares whether the number of peaks is greater or less than a specific value (indicated by threshold-hold TH in the figure), and if it is greater than the THO value, it is output as audio through the control interface circuit 46. Note that the comparison circuit 43
, the adder 44, and the comparator 45 form the determination section 3 in FIG.

FIG. 5 is a diagram showing the configuration of still another specific example of the automatic disabler of FIG. 1 according to the present invention. The criterion for this device is based on the fact that since a non-speech signal is a modulated positive wave, there are fewer harmonics than in the case of speech. In order to detect this state, the level difference between the maximum peak and the second peak of the frequency analysis circuit is measured. In FIG. 5, the signal inputted by two people is divided into each band by FIL 51, and the signal level measuring device 5
2. Measure the signal level. However, when this measured value is less than a predetermined value, it is set to O for stability of operation. Using this measured value, the comparator circuit 53
The peak of the spectrum is determined by , and points other than the peak are set to 1) 0 according to the gate 54 .

The maximum level second level detection circuit compares the input peak signal levels and outputs the maximum level and the second level together. A divider 56 calculates the level ratio between the maximum level and the second level, and a 1 comparison circuit 57 compares whether this value is greater than a predetermined value. When this level ratio exceeds a predetermined value TH, it is outputted through the interface circuit 58 as a non-audio signal. As can be seen from the above description, 51 and 52 form the frequency analysis section 2 in FIG. 1, and 52 to 57 form the determination section 3. .

As explained above, 1. According to the present invention, it is possible to detect non-voice communication that does not send out a conditioning tone.
It is possible to prevent data errors from occurring due to the operation of the echo control device.

Note that the apparatus shown in FIG. 3 and the apparatus shown in FIG. 4 can be used in combination. In other words, if the output of the frequency analysis section 3 is applied in parallel to the determination section shown in FIG. 3 and the IJ constant section shown in FIG. You can increase crabs. In this case, since the frequency analysis section can be shared, it is economical to simply install the two devices in parallel.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the basic configuration of an automatic day saver for a reverberation control device according to the present invention. FIG. 2 is a diagram showing the normal shape of the spectrum of the sound and the modulated wave; FIGS. It is a figure showing. Explanation of symbols: 1 is the input terminal, 2 is the frequency analysis section. 6 is a determination unit, 4 is a control circuit, 5 is an output terminal, 21 to 24
is a filter, 25 to 28 are signal level measurement circuits, 31 is a comparison circuit, 321d is a judgment circuit, 41 is a filter, 42 is a signal level measurement circuit, 4'5 is a comparison circuit, 44u adder, 45 is a ratio 1 articulator, 55 is a maximum level second level detection circuit, and 56 is a divider. TH has Slensey Hole Toys. I! aZ! :=

Claims (1)

[Claims] (1) For an echo control device that is connected to an echo control device used to eliminate the influence of echoes generated on a communication line and stops the operation of the echo control device when one communication is a non-voice communication. In the automatic disabler, a determination circuit determines whether the input signal is an audio signal or a non-audio signal based on the output of the frequency analysis circuit and the frequency analysis circuit. An automatic disassembler for a echo control device, characterized by comprising a control circuit that uses the results to control the operation of the echo control device. (2. In the automatic disabler for the echo control device-1 according to claim 1, the judgment and the fixed path are based on the signal level at each frequency, number, and $ that is the output of the frequency analysis circuit. □ An automatic disabler for a reverberation control device according to claim 1, which is a judgment circuit for making a determination, and is characterized by the following characteristics: The determination circuit is characterized in that it is a determination circuit that makes a determination based on the number of frequency bands in which the levels of the preceding and succeeding frequency bands are lower among the signal levels in each frequency band that are output from the frequency analysis circuit. (4) In the automatic de-sailor for an echo control device according to claim 1, the determination circuit determines whether the signal level in each frequency band output from the frequency analysis circuit is An automatic display device for an echo control device, characterized in that it is equipped with a determination circuit that makes a determination based on the level of the difference between the maximum level point and the next level point in the level of the frequency band where the level of the frequency band is low.・Isra.